In this paper anti-spam filtering is presented as a cumbersome service, as opposed to a software product perspective. The huge human effort for setting up, adaptation, maintenance, and tuning of filters for spam detection in anti-spam systems is explained. Choosing the best importance scores for the spam filters is essential for the accuracy of any rules based anti-spam system, and is also one of the biggest challenges in this research area. Optimal filters score settings for Apache SpamAssassin project (the most widely adopted anti-spam open-source software) is addressed. In addition to a survey done on single/multi-objective optimization research in this area, we also present a study for filters score setting using multiobjective optimization based on two most representative evolutionary algorithms, NSGA II and SPEA2. Problem description, simulation and results analysis is done for SpamAssassin public mail corpus which is widely used for benchmarking purposes.